Mauro Turner
ES Let's talk
All services

Repetitive work, off your team's calendar

Intelligent Process Automation. Every figure copied by hand between the ERP, SAP and a spreadsheet is lost time and an error waiting to surface. I design pipelines and workflows with Airflow, Power Automate and n8n that ingest data, connect systems and remove administrative bottlenecks.

When it makes sense

  • Someone spends hours each week copying data between systems.
  • A process stops when one person goes on holiday.
  • Errors only surface once a customer complains.

What I deliver

  • Map of the process and its bottlenecks
  • Ingestion pipelines orchestrated in Airflow
  • Workflows in Power Automate and n8n
  • Integrations with ERPs, SAP and APIs
  • Alerts and a log of every run
  • Operating guide for your team

What I work with

  • Airflow
  • Power Automate
  • n8n
  • SAP
  • ERPs
  • APIs

Your day starts at 9:00. The system started earlier.

This is what a morning can look like with the processes automated. By the time you arrive, the repetitive work is done.

Illustrative run

  1. Data extracted

    Sales from SAP, orders from an API and the spreadsheets are validated and loaded into the model.

  2. Reports built

    The sales and stock reports put themselves together, with yesterday’s numbers.

  3. Emails sent

    Each team gets its summary, and collections gets the detail of what is overdue.

  4. Invoices read

    A model reads the suppliers’ PDFs and enters the amounts in the ERP.

  5. Questions answered

    An agent answers the common questions and passes the rest to someone on your team.

  6. Alerts flagged

    An unusual order and a low stock are marked for review.

  7. Dashboard up to date

    The summary of the run, ready for the first meeting.

    Hours saved
    64 h
    Runs without error
    98.3 %
    Cycle time
    3 min
    Failures
    1

    Synthetic data

  8. You walk in.

    What is left on your agenda is what matters: deciding.

What it looks like

Examples built on synthetic data to show the shape of each deliverable. None belongs to a client.

Orchestrating a daily ingest

Every step is logged. If one fails, it retries and tells you.

Synthetic data

Hours saved
64 h
per month
Runs without error
98.3 %
last 12 weeks
Cycle time
3 min
before, 5 h by hand
Failures
1
last 12 weeks
Task Timeline Status Duration
Extract sales from SAP Complete 42 s
Read orders from the API Complete 18 s
Validate and deduplicate Complete after one retry 1 min 06 s
Load into the model Complete 27 s
Refresh dashboard and notify Complete 9 s

Runs over the last 12 weeks

View the data as a table
Runs over the last 12 weeks
Week MonTueWedThuFriSatSun
1 No issuesNo issuesNo issuesNo issuesNo issuesNo runNo run
2 No issuesNo issuesRetriedNo issuesNo issuesNo runNo run
3 No issuesNo issuesNo issuesNo issuesNo issuesNo runNo run
4 No issuesNo issuesNo issuesNo issuesRetriedNo runNo run
5 No issuesNo issuesNo issuesNo issuesNo issuesNo runNo run
6 No issuesNo issuesNo issuesNo issuesNo issuesNo runNo run
7 No issuesFailedNo issuesNo issuesNo issuesNo runNo run
8 No issuesNo issuesNo issuesNo issuesNo issuesNo runNo run
9 No issuesRetriedNo issuesNo issuesNo issuesNo runNo run
10 No issuesNo issuesNo issuesNo issuesNo issuesNo runNo run
11 No issuesNo issuesNo issuesNo issuesNo issuesNo runNo run
12 No issuesNo issuesNo issuesNo issuesNo issuesNo runNo run

Does your process stall on this line?

Two lines on what runs late are enough to start.

Tell me about it